Status epilepticus is associated with increased morbidity and mortality. Early and rapid escalation of treatment is critical to preventing neuronal injury. Treatment guidelines recommend initial antiepileptic therapy at 5 minutes and rapid escalation to second-phase therapy if seizure persists. A recent prospective multicenter study and preintervention review at our institution confirmed delayed treatment in hospitalized children.
Our taskforce designed a quality improvement strategy to rapidly treat status epilepticus in the emergency department and inpatient units. We aimed to decrease initial phase antiepileptic benzodiazepine (BZD) administration time for acute seizure from 7 (preintervention) to ≤5 minutes and to decrease non-BZD (phenobarbital, phenytoin, levetiracetam, valproic acid) medication order to administration time from 28 (preintervention) to <10 minutes for patients with persistent seizure requiring urgent pharmaceutical therapy.
A multidisciplinary team developed a key driver diagram (Fig. 1) to identify opportunities to improve efficiency. Interventions targeted team communication and medication delivery. All Pyxis were stocked with intranasal/buccal BZD available on “nursing override” for easy access and immediate administration. Because non-BZDs require pharmacy compounding and the pharmacy receives many STAT orders with competing priorities, our team developed a hospital-wide Seizure Rescue process using patient room staff terminals to mobilize pharmacy to the bedside for immediate medication administration (Fig. 2). We educated care teams about rapid treatment goals and trained nurses to press the Seizure Rescue button for seizures requiring urgent therapy. This process activated the Seizure Rescue–assigned pharmacist to be present at the bedside with a backpack stocked with non-BZDs ready to compound. We reviewed the medical records of all patients with Seizure Rescue activations.
Of the 105 seizures during the postintervention period (July 2016 to October 2017), 61 cases (49%) required urgent therapy. Median time from seizure onset to initial BZD therapy decreased from 7 (preintervention) to 2 minutes (postintervention). Median time from order to administration of non-BZDs decreased from 28 (preintervention) to 11 minutes (postintervention) (Fig. 3).
By leveraging existing patient room technology infrastructure to mobilize pharmacy to the bedside, we expedited non-BZD administration by 60%. Further, the rapid response Seizure Rescue process may have created an increased sense of overall clinical urgency to treat seizures, helping to expedite initial BZD administration by 70%. This consequently may have prevented progression to status epilepticus, the need for second-phase therapy, and preventable neuronal injury—all without requiring the additional resources of a code team.